Quality Control is a critical pillar of medical device manufacturing. After all, if products do not meet the highest standards of safety and functionality, people could get misdiagnosed, hurt or even die. Every device must adhere to exact specifications and regulations; even minor deviations can have significant repercussions.
Using an example like optimizing a component for MRI, we’ll highlight how Minitab Statistical Software can help you quickly identify sources of risk in your production process and address them, saving time, money and potentially lives as well.
How is coil resistance measured?
In medical device manufacturing, coil resistance, especially in Magnetic Resonance Imaging (MRI) systems, denotes how easily electrical current flows through the coils within these machines. An ohm is the unit of measurement for electrical resistance, representing the resistance encountered by an electrical current when passing through a material. It’s vital as it directly impacts the magnetic fields’ quality, crucial for clear and accurate diagnostic images. By managing coil resistance levels, manufacturers ensure better image clarity, regulatory compliance, and safety, which enhances diagnostic accuracy and patient well-being.
Coil resistance measurements typically fall within specific ranges in MRI manufacturing to ensure performance and safety. For instance, in our case, transmit coils, which generate radiofrequency pulses, often exhibit higher resistance readings. Understanding these ranges is crucial since they affect magnetic field efficiency, which is critical for accurate imaging. Precise coil resistance measurements within this range maintain MRI system quality and reliability, with deviations prompting further investigation to uphold stringent standards.
We pulled sample data from a manufacturer of these coils to track if the ohm ratings were consistent and, if not, to determine which factors led to a statistically significant variation in ohm strength.
If you work as an operator, manager, or director at an organization that manufactures any type of product, this post is for you.
See What’s New with Minitab 22. Watch Our On-demand Webinar.
What do control charts show about our data?
To begin, we collected data from 50 manufacturing runs and determined the mean resistance reading from each run. Then, we put the data into Minitab Statistical Software (this process can be done automatically in real time with Real-Time SPC). We then prompted Minitab to produce an I-MR Chart of the coil resistance measurement (measured in ohms):
This chart showed us concerning data points where the measurement was much higher than it should have been on the individual value chart. The team was now able to point to three specific times when the average coil resistance was much higher than it should have been, or out of control.
What factors could cause the variability?
The team then decided that they needed to brainstorm to determine possible variables that could potentially cause variability in MRI coil resistance. To do this, the team got together and discussed possible factors that could potentially lead to variability. They determined that the variability was likely caused by one of four general attributes: issues related to the raw materials, issues related to the process, design problems, or human error. They then determined a few causes under each factor that could lead to significant variability.
They then went to Minitab Workspace and created a Fishbone Diagram to visualize the potential variables:
They now had a polished, clear diagram to present to leadership outlining possible causes of the variability in coil resistance that the control charts indicated.
Try Minitab Workspace for Free.
What about other factors?
As the team looked further into other sources of variation to determine what might be causing some measurements to fall out of the expected range, they decided to first focus on the human factor and possible operator error.
There were six operators total, and each line had two separate operators working on them at the time of the data collection. The team deployed a one-way ANOVA test. In simple terms, one-way ANOVA shows whether there are significant differences in the means of three or more independent groups. Here were their results:
Visually, operator O had far higher readings than all the other operators. And with a P-Value of .002, it was fair to say that this discrepancy was indeed statistically significant.
What’s the value of using data in medical device manufacturing?
In this instance, thousands of dollars were saved. By investigating and removing a significant source of variation in their process by retraining operator O, they were able to make the process more capable of meeting specifications. With some remedial training and additional support and oversight, the operator improved and lowered the ohms reading of the coils he was responsible for producing.
Additionally, being able to pinpoint a specific operator was key. Instead of retraining the whole staff—a time intensive and costly endeavor—efforts could be directed to the one person who needed them the most. And bias was removed; the p-value proved that the difference between operator O and the other operators was indeed statistically significant and could not be explained away by chance.
Minitab can also be used for:
- Efficient handling of extensive datasets for comprehensive analysis.
- Conducting complex statistical analyses to uncover insights and trends.
- Generating customizable reports tailored to specific regulatory requirements and quality standards.
Getting started with Minitab is easy, regardless of your role. You don’t need to be a statistician, and Minitab has a dedicated team of professionals to help with deployment or training.
Be the first to comment